ubi: fastmap: Fix a series of wear leveling problems
Hi,
I notice a bug report with proposed fixes on Bugzilla [1]. Quoting from it (only the first problem that is quoted):
Quoted 27 lines
Problem 1: large erase counter for single fastmap data PEB
Config:
x86_64qemu
flash: nandsimCONFIG_MTD_UBI_WL_THRESHOLD=128CONFIG_MTD_UBI_FASTMAP=y
ubi->beb_rsvd_pebs=0Running fsstress on ubifs for 3h(fastmap data PEB has large erase counter than others):
=========================================================
from to count min avg max
---------------------------------------------------------0 .. 9: 0 0 0 0 10 .. 99: 532 84 92 99 100 .. 999: 15787 100 147 229 1000 .. 9999: 64 4699 4765 4826 10000 .. 99999: 0 0 0 0 100000 .. inf: 1 272935 272935 272935---------------------------------------------------------
Total : 16384 84 180 272935
PEB 8031(ec=272935) is always taken for fastmap data.After fix, running fsstress on ubifs for 12h(no pool reservation), no individual peb has big erase counter:
=========================================================
from to count min avg max
---------------------------------------------------------
Quoted 8 lines
0 .. 9: 0 0 0 0 10 .. 99: 0 0 0 0 100 .. 999: 16320 609 642 705 1000 .. 9999: 0 0 0 0 10000 .. 99999: 64 18176 18234 18303 100000 .. inf: 0 0 0 0---------------------------------------------------------
Total : 16384 609 710 18303
See Bugzilla for the full thread (with other problems mentioned) and attached patch series that fixes them.
Zhihao: I asked you on BZ to send your patches to linux-mtd list, but you didn't respond there. Would you like to send them for review?
Thanks.
[1]: https://bugzilla.kernel.org/show_bug.cgi?id=217787
Signature · 5 lines
An old man doll... just what I always wanted! - Clara ______________________________________________________ Linux MTD discussion mailing list http://lists.infradead.org/mailman/listinfo/linux-mtd/
在 2023/8/16 17:56, Bagas Sanjaya 写道:
Hi,
Quoted 41 lines
Hi,
I notice a bug report with proposed fixes on Bugzilla [1]. Quoting from it (only the first problem that is quoted):
Quoted 27 lines
Problem 1: large erase counter for single fastmap data PEB
Config:
x86_64qemu
flash: nandsimCONFIG_MTD_UBI_WL_THRESHOLD=128CONFIG_MTD_UBI_FASTMAP=y
ubi->beb_rsvd_pebs=0Running fsstress on ubifs for 3h(fastmap data PEB has large erase counter than others):
=========================================================
from to count min avg max
---------------------------------------------------------0 .. 9: 0 0 0 0 10 .. 99: 532 84 92 99 100 .. 999: 15787 100 147 229 1000 .. 9999: 64 4699 4765 4826 10000 .. 99999: 0 0 0 0 100000 .. inf: 1 272935 272935 272935---------------------------------------------------------
Total : 16384 84 180 272935
PEB 8031(ec=272935) is always taken for fastmap data.After fix, running fsstress on ubifs for 12h(no pool reservation), no individual peb has big erase counter:
=========================================================
from to count min avg max
---------------------------------------------------------Quoted 8 lines
0 .. 9: 0 0 0 0 10 .. 99: 0 0 0 0 100 .. 999: 16320 609 642 705 1000 .. 9999: 0 0 0 0 10000 .. 99999: 64 18176 18234 18303 100000 .. inf: 0 0 0 0---------------------------------------------------------
Total : 16384 609 710 18303
Quoted 7 lines
See Bugzilla for the full thread (with other problems mentioned) and attached patch series that fixes them.
Zhihao: I asked you on BZ to send your patches to linux-mtd list, but you didn't respond there. Would you like to send them for review?
Sorry for the delayed response. Yes, I have sent the fix patches in https://patchwork.ozlabs.org/project/linux-mtd/list/?series=368534.
Any suggestions are welcomed.
Quoted 4 lines
Thanks.
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/